[ci]: fix fork PR checkout — pass PR_NUMBER through Buildkite to Modal for ref-based checkout

.buildkite/pipeline.yml

@@ -5,32 +5,17 @@ env:
 notify:
   - github_commit_status:
       context: "fastcheck-passed"
-    if: build.env("TEST_SCOPE") == "fastcheck" || build.env("TEST_SCOPE") == null
   - github_commit_status:
       context: "full-suite-passed"
-    if: build.env("TEST_SCOPE") == "full"
+    if: build.env("FULL_SUITE") == "true"
 
 steps:
-  # ============================================================
-    - label: ":dart: Direct Test (${TEST_TYPE})"
-      if: build.env("TEST_SCOPE") == "direct"
-      command: "timeout 90m .buildkite/scripts/pr_test.sh"
-      retry:
-        automatic:
-          - exit_status: 128
-            limit: 3
-          - exit_status: -1
-            limit: 2
-      agents:
-        queue: "default"
-
   # ============================================================
   # Fastcheck: Runs on every PR (~10-15 min parallel)
   # Core component validation: encoders, VAEs, transformers,
   # CUDA kernels, and unit tests.
   # ============================================================
     - label: "Trigger Fastcheck"
-      if: build.env("TEST_SCOPE") == "fastcheck" || build.env("TEST_SCOPE") == null
       retry:
         automatic:
           - exit_status: 128
@@ -108,14 +93,14 @@ steps:
                   queue: "default"
 
   # ============================================================
-  # Full Suite: Runs when TEST_SCOPE=full
+  # Full Suite: Runs when FULL_SUITE=true
   # Triggered by adding the 'ready' label (via ci-trigger-full-suite.yml)
   # or on-demand via /test full slash command.
   # Includes integration tests, SSIM regression, training pipelines,
   # and performance benchmarks.
   # ============================================================
     - label: "Trigger Full Suite"
-      if: build.env("TEST_SCOPE") == "full"
+      if: build.env("FULL_SUITE") == "true"
       retry:
         automatic:
           - exit_status: 128

.github/mergify.yml

@@ -84,12 +84,12 @@ pull_request_rules:
   - name: "label scope: training"
     conditions:
       - or:
-          - files~=^fastvideo/train/
-          - files~=^fastvideo/training/
-          - files~=^fastvideo/distillation/
-          - files~=^examples/train/
-          - files~=^examples/training/
-          - files~=^examples/distill/
+        - files~=^fastvideo/train/
+        - files~=^fastvideo/training/
+        - files~=^fastvideo/distillation/
+        - files~=^examples/train/
+        - files~=^examples/training/
+        - files~=^examples/distill/
       - -closed
     actions:
       label:
@@ -98,14 +98,14 @@ pull_request_rules:
   - name: "label scope: inference"
     conditions:
       - or:
-          - files~=^fastvideo/pipelines/basic/
-          - files~=^fastvideo/pipelines/stages/
-          - files~=^fastvideo/pipelines/samplers/
-          - files~=^fastvideo/entrypoints/
-          - files~=^fastvideo/worker/
-          - files~=^fastvideo/configs/sample/
-          - files~=^fastvideo/configs/pipelines/
-          - files~=^examples/inference/
+        - files~=^fastvideo/pipelines/basic/
+        - files~=^fastvideo/pipelines/stages/
+        - files~=^fastvideo/pipelines/samplers/
+        - files~=^fastvideo/entrypoints/
+        - files~=^fastvideo/worker/
+        - files~=^fastvideo/configs/sample/
+        - files~=^fastvideo/configs/pipelines/
+        - files~=^examples/inference/
       - -closed
     actions:
       label:
@@ -122,8 +122,8 @@ pull_request_rules:
   - name: "label scope: kernel"
     conditions:
       - or:
-          - files~=^fastvideo-kernel/
-          - files~=^csrc/
+        - files~=^fastvideo-kernel/
+        - files~=^csrc/
       - -closed
     actions:
       label:
@@ -132,9 +132,9 @@ pull_request_rules:
   - name: "label scope: data"
     conditions:
       - or:
-          - files~=^fastvideo/dataset/
-          - files~=^fastvideo/pipelines/preprocess/
-          - files~=^examples/preprocessing/
+        - files~=^fastvideo/dataset/
+        - files~=^fastvideo/pipelines/preprocess/
+        - files~=^examples/preprocessing/
       - -closed
     actions:
       label:
@@ -143,10 +143,10 @@ pull_request_rules:
   - name: "label scope: infra"
     conditions:
       - or:
-          - files~=^\.github/
-          - files~=^\.buildkite/
-          - files~=^fastvideo/tests/
-          - files~=^docker/
+        - files~=^\.github/
+        - files~=^\.buildkite/
+        - files~=^fastvideo/tests/
+        - files~=^docker/
       - -closed
     actions:
       label:
@@ -179,9 +179,9 @@ pull_request_rules:
   - name: "label scope: model"
     conditions:
       - or:
-          - files~=^fastvideo/models/
-          - files~=^fastvideo/layers/
-          - files~=^fastvideo/configs/models/
+        - files~=^fastvideo/models/
+        - files~=^fastvideo/layers/
+        - files~=^fastvideo/configs/models/
       - -closed
     actions:
       label:

.github/workflows/ci-precommit.yml

@@ -14,7 +14,7 @@ permissions:
 
 jobs:
   pre-commit:
-    if: github.event_name == 'workflow_call' || github.event.pull_request.draft != true
+    if: github.event.pull_request.draft != true
     runs-on: ubuntu-latest
     steps:
     - uses: actions/checkout@v4

.github/workflows/ci-slash-commands.yml

@@ -54,7 +54,6 @@ jobs:
     runs-on: ubuntu-latest
     outputs:
       test_type: ${{ steps.parse.outputs.test_type }}
-      test_scope: ${{ steps.parse.outputs.test_scope }}
       full_suite: ${{ steps.parse.outputs.full_suite }}
       pr_sha: ${{ steps.pr.outputs.sha }}
       pr_branch: ${{ steps.pr.outputs.branch }}
@@ -86,7 +85,7 @@ jobs:
           set -euo pipefail
           TEST_NAME=$(echo "$COMMENT" | grep -oP '(?<=/test\s)\S+' | head -1 || true)
 
-          VALID="encoder vae transformer kernel unit ssim training lora-inference lora-training distillation self-forcing vsa vmoba performance api full fastcheck pre-commit"
+          VALID="encoder vae transformer kernel unit ssim training lora-inference lora-training distillation self-forcing vsa vmoba performance api full fastcheck"
           if [ -z "$TEST_NAME" ] || ! echo "$VALID" | grep -qw "$TEST_NAME"; then
             echo "Unknown test: '$TEST_NAME'. Valid: $VALID"
             exit 1
@@ -103,29 +102,14 @@ jobs:
           )
 
           if [ "$TEST_NAME" = "full" ]; then
-            {
-              echo "test_type=all"
-              echo "test_scope=full"
-              echo "full_suite=true"
-            } >> "$GITHUB_OUTPUT"
+            echo "test_type=all" >> "$GITHUB_OUTPUT"
+            echo "full_suite=true" >> "$GITHUB_OUTPUT"
           elif [ "$TEST_NAME" = "fastcheck" ]; then
-            {
-              echo "test_type=fastcheck"
-              echo "test_scope=fastcheck"
-              echo "full_suite=false"
-            } >> "$GITHUB_OUTPUT"
-          elif [ "$TEST_NAME" = "pre-commit" ]; then
-            {
-              echo "test_type="
-              echo "test_scope=precommit"
-              echo "full_suite=false"
-            } >> "$GITHUB_OUTPUT"
+            echo "test_type=fastcheck" >> "$GITHUB_OUTPUT"
+            echo "full_suite=false" >> "$GITHUB_OUTPUT"
           else
-            {
-              echo "test_type=${MAP[$TEST_NAME]}"
-              echo "test_scope=direct"
-              echo "full_suite=false"
-            } >> "$GITHUB_OUTPUT"
+            echo "test_type=${MAP[$TEST_NAME]}" >> "$GITHUB_OUTPUT"
+            echo "full_suite=true" >> "$GITHUB_OUTPUT"
           fi
 
       - name: Get PR details
@@ -142,35 +126,6 @@ jobs:
             core.setOutput('sha', pr.head.sha);
             core.setOutput('branch', pr.head.ref);
 
-  pre-commit:
-    needs: parse-command
-    if: >-
-      needs.parse-command.outputs.has_write == 'true'
-      && needs.parse-command.outputs.test_scope == 'precommit'
-    uses: ./.github/workflows/ci-precommit.yml
-
-  post-precommit-status:
-    needs: [parse-command, pre-commit]
-    if: always() && needs.parse-command.outputs.test_scope == 'precommit'
-    runs-on: ubuntu-latest
-    steps:
-      - name: Post commit status
-        uses: actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea # v7.0.1
-        env:
-          PR_SHA: ${{ needs.parse-command.outputs.pr_sha }}
-          RESULT: ${{ needs.pre-commit.result }}
-        with:
-          script: |
-            const state = process.env.RESULT === 'success' ? 'success' : 'failure';
-            await github.rest.repos.createCommitStatus({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              sha: process.env.PR_SHA,
-              state,
-              context: 'pre-commit',
-              description: `Triggered via /test pre-commit (${state})`,
-            });
-
   trigger-buildkite:
     needs: parse-command
     if: >-
@@ -190,41 +145,14 @@ jobs:
             });
 
       - name: Trigger Buildkite
-        env:
-          BUILDKITE_API_TOKEN: ${{ secrets.BUILDKITE_API_TOKEN }}
-          PR_SHA: ${{ needs.parse-command.outputs.pr_sha }}
-          PR_BRANCH: ${{ needs.parse-command.outputs.pr_branch }}
-          PR_NUMBER: ${{ github.event.issue.number }}
-          TEST_SCOPE: ${{ needs.parse-command.outputs.test_scope }}
-          FULL_SUITE: ${{ needs.parse-command.outputs.full_suite }}
-          TEST_TYPE: ${{ needs.parse-command.outputs.test_type }}
-          BK_ORG: ${{ vars.BUILDKITE_ORG_SLUG }}
-          BK_PIPELINE: ${{ vars.BUILDKITE_PIPELINE_SLUG }}
-        run: |
-          curl -sS --fail-with-body -X POST \
-            "https://api.buildkite.com/v2/organizations/${BK_ORG}/pipelines/${BK_PIPELINE}/builds" \
-            -H "Authorization: Bearer $BUILDKITE_API_TOKEN" \
-            -H "Content-Type: application/json" \
-            --data-raw "$(jq -n \
-              --arg commit "$PR_SHA" \
-              --arg branch "$PR_BRANCH" \
-              --arg message "/test ${TEST_TYPE} on PR #${PR_NUMBER}" \
-              --argjson pr_id "$PR_NUMBER" \
-              --arg test_scope "$TEST_SCOPE" \
-              --arg full_suite "$FULL_SUITE" \
-              --arg test_type "$TEST_TYPE" \
-              --arg pr_number "$PR_NUMBER" \
-              '{
-                commit: $commit,
-                branch: $branch,
-                message: $message,
-                ignore_pipeline_branch_filters: true,
-                pull_request_id: $pr_id,
-                pull_request_base_branch: "main",
-                env: {
-                  TEST_SCOPE: $test_scope,
-                  FULL_SUITE: $full_suite,
-                  TEST_TYPE: $test_type,
-                  PR_NUMBER: $pr_number
-                }
-              }')"
+        uses: buildkite/trigger-pipeline-action@v2.4.1
+        with:
+          buildkite_api_access_token: ${{ secrets.BUILDKITE_API_TOKEN }}
+          pipeline: "${{ vars.BUILDKITE_ORG_SLUG }}/${{ vars.BUILDKITE_PIPELINE_SLUG }}"
+          branch: "${{ needs.parse-command.outputs.pr_branch }}"
+          commit: "${{ needs.parse-command.outputs.pr_sha }}"
+          message: "/test ${{ needs.parse-command.outputs.test_type }} on PR #${{ github.event.issue.number }}"
+          build_env_vars: '{"FULL_SUITE": "${{ needs.parse-command.outputs.full_suite }}", "TEST_TYPE": "${{ needs.parse-command.outputs.test_type }}", "PR_NUMBER": "${{ github.event.issue.number }}"}'
+          send_pull_request: "true"
+          pull_request_id: "${{ github.event.issue.number }}"
+          pull_request_base_branch: "main"

.github/workflows/ci-trigger-full-suite.yml

@@ -39,10 +39,10 @@ jobs:
           BUILDKITE_API_TOKEN: ${{ secrets.BUILDKITE_API_TOKEN }}
           PR_BRANCH: ${{ github.event.pull_request.head.ref }}
         run: |
-          # Find running builds for this branch with TEST_SCOPE=full and cancel them
+          # Find running builds for this branch with FULL_SUITE=true and cancel them
           builds=$(curl -sS -H "Authorization: Bearer $BUILDKITE_API_TOKEN" \
             "https://api.buildkite.com/v2/organizations/${{ vars.BUILDKITE_ORG_SLUG }}/pipelines/${{ vars.BUILDKITE_PIPELINE_SLUG }}/builds?branch=${PR_BRANCH}&state=running,scheduled" \
-            | jq -r '.[] | select(.env.TEST_SCOPE == "full") | .number')
+            | jq -r '.[] | select(.env.FULL_SUITE == "true") | .number')
           for build_num in $builds; do
             echo "Cancelling Buildkite build #$build_num"
             curl -sS -X PUT -H "Authorization: Bearer $BUILDKITE_API_TOKEN" \
@@ -51,33 +51,14 @@ jobs:
 
       - name: Trigger Buildkite Full Suite
         if: steps.check.outputs.has_ready == 'true'
-        env:
-          BUILDKITE_API_TOKEN: ${{ secrets.BUILDKITE_API_TOKEN }}
-          PR_SHA: ${{ github.event.pull_request.head.sha }}
-          PR_BRANCH: ${{ github.event.pull_request.head.ref }}
-          PR_NUMBER: ${{ github.event.pull_request.number }}
-          BK_ORG: ${{ vars.BUILDKITE_ORG_SLUG }}
-          BK_PIPELINE: ${{ vars.BUILDKITE_PIPELINE_SLUG }}
-        run: |
-          curl -sS --fail-with-body -X POST \
-            "https://api.buildkite.com/v2/organizations/${BK_ORG}/pipelines/${BK_PIPELINE}/builds" \
-            -H "Authorization: Bearer $BUILDKITE_API_TOKEN" \
-            -H "Content-Type: application/json" \
-            --data-raw "$(jq -n \
-              --arg commit "$PR_SHA" \
-              --arg branch "$PR_BRANCH" \
-              --arg message "Full Suite for PR #${PR_NUMBER}" \
-              --argjson pr_id "$PR_NUMBER" \
-              '{
-                commit: $commit,
-                branch: $branch,
-                message: $message,
-                ignore_pipeline_branch_filters: true,
-                pull_request_id: $pr_id,
-                pull_request_base_branch: "main",
-                env: {
-                  TEST_SCOPE: "full",
-                  FULL_SUITE: "true",
-                  PR_NUMBER: ($pr_id | tostring)
-                }
-              }')"
+        uses: buildkite/trigger-pipeline-action@v2.4.1
+        with:
+          buildkite_api_access_token: ${{ secrets.BUILDKITE_API_TOKEN }}
+          pipeline: "${{ vars.BUILDKITE_ORG_SLUG }}/${{ vars.BUILDKITE_PIPELINE_SLUG }}"
+          branch: "${{ github.event.pull_request.head.ref }}"
+          commit: "${{ github.event.pull_request.head.sha }}"
+          message: "Full Suite for PR #${{ github.event.pull_request.number }}"
+          build_env_vars: '{"FULL_SUITE": "true", "PR_NUMBER": "${{ github.event.pull_request.number }}"}'
+          send_pull_request: "true"
+          pull_request_id: "${{ github.event.pull_request.number }}"
+          pull_request_base_branch: "main"

docs/contributing/ci_architecture.md

@@ -277,8 +277,8 @@ Triggers a specific Buildkite test or suite on the current PR branch.
 | `/test vmoba` | VMoBA inference tests | `inference_vmoba` |
 | `/test performance` | Performance benchmarks | `performance` |
 | `/test api` | API server integration tests | `api_server` |
-| `/test full` | Entire Full Suite | all (with `TEST_SCOPE=full`) |
-| `/test fastcheck` | Entire Fastcheck suite | fastcheck (with `TEST_SCOPE=fastcheck`) |
+| `/test full` | Entire Full Suite | all (with `FULL_SUITE=true`) |
+| `/test fastcheck` | Entire Fastcheck suite | fastcheck (with `FULL_SUITE=false`) |
 
 ---
 

fastvideo/tests/modal/pr_test.py

@@ -9,24 +9,22 @@ image_version = os.getenv("IMAGE_VERSION")
 image_tag = f"ghcr.io/hao-ai-lab/fastvideo/fastvideo-dev:{image_version}"
 print(f"Using image: {image_tag}")
 
-image = (modal.Image.from_registry(
-    image_tag, add_python="3.12"
-).run_commands("rm -rf /FastVideo").apt_install(
-    "cmake", "pkg-config", "build-essential", "curl", "libssl-dev", "ffmpeg"
-).run_commands(
-    "curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --default-toolchain stable"
-).run_commands("echo 'source ~/.cargo/env' >> ~/.bashrc").env({
-    "PATH":
-    "/root/.cargo/bin:$PATH",
-    "BUILDKITE_REPO":
-    os.environ.get("BUILDKITE_REPO", ""),
-    "BUILDKITE_COMMIT":
-    os.environ.get("BUILDKITE_COMMIT", ""),
-    "BUILDKITE_PULL_REQUEST":
-    os.environ.get("BUILDKITE_PULL_REQUEST", ""),
-    "IMAGE_VERSION":
-    os.environ.get("IMAGE_VERSION", ""),
-}))
+image = (
+    modal.Image.from_registry(image_tag, add_python="3.12")
+    .run_commands("rm -rf /FastVideo")
+    .apt_install("cmake", "pkg-config", "build-essential", "curl", "libssl-dev", "ffmpeg")
+    .run_commands("curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y --default-toolchain stable")
+    .run_commands("echo 'source ~/.cargo/env' >> ~/.bashrc")
+    .env(
+        {
+            "PATH": "/root/.cargo/bin:$PATH",
+            "BUILDKITE_REPO": os.environ.get("BUILDKITE_REPO", ""),
+            "BUILDKITE_COMMIT": os.environ.get("BUILDKITE_COMMIT", ""),
+            "BUILDKITE_PULL_REQUEST": os.environ.get("BUILDKITE_PULL_REQUEST", ""),
+            "IMAGE_VERSION": os.environ.get("IMAGE_VERSION", ""),
+        }
+    )
+)
 
 
 def run_test(pytest_command: str):
@@ -41,8 +39,7 @@ def run_test(pytest_command: str):
 
     print(f"Cloning repository: {git_repo}")
     print(f"Target commit: {git_commit}")
-    if pr_number:
-        print(f"PR number: {pr_number}")
+    print(f"PR number: {pr_number}")
 
     # For PRs (including forks), use GitHub's PR refs to get the correct commit
     if pr_number and pr_number != "false":
@@ -66,95 +63,84 @@ def run_test(pytest_command: str):
     {pytest_command}
     """
 
-    result = subprocess.run(["/bin/bash", "-c", command],
-                            stdout=sys.stdout,
-                            stderr=sys.stderr,
-                            check=False)
+    result = subprocess.run(["/bin/bash", "-c", command], stdout=sys.stdout, stderr=sys.stderr, check=False)
 
     sys.exit(result.returncode)
 
 
-@app.function(gpu="H100:1",
-              image=image,
-              timeout=1200,
-              secrets=[
-                  modal.Secret.from_dict(
-                      {"HF_API_KEY": os.environ.get("HF_API_KEY", "")})
-              ],
-              volumes={"/root/data": model_vol})
+@app.function(
+    gpu="H100:1",
+    image=image,
+    timeout=1200,
+    secrets=[modal.Secret.from_dict({"HF_API_KEY": os.environ.get("HF_API_KEY", "")})],
+    volumes={"/root/data": model_vol},
+)
 def run_encoder_tests():
     run_test(
         "export HF_HOME='/root/data/.cache' && hf auth login --token $HF_API_KEY && pytest ./fastvideo/tests/encoders -vs"
     )
 
 
-@app.function(gpu="L40S:1",
-              image=image,
-              timeout=1200,
-              secrets=[
-                  modal.Secret.from_dict(
-                      {"HF_API_KEY": os.environ.get("HF_API_KEY", "")})
-              ],
-              volumes={"/root/data": model_vol})
+@app.function(
+    gpu="L40S:1",
+    image=image,
+    timeout=1200,
+    secrets=[modal.Secret.from_dict({"HF_API_KEY": os.environ.get("HF_API_KEY", "")})],
+    volumes={"/root/data": model_vol},
+)
 def run_vae_tests():
     run_test(
         "export HF_HOME='/root/data/.cache' && hf auth login --token $HF_API_KEY && pytest ./fastvideo/tests/vaes -vs"
     )
 
 
-@app.function(gpu="L40S:1",
-              image=image,
-              timeout=900,
-              secrets=[
-                  modal.Secret.from_dict(
-                      {"HF_API_KEY": os.environ.get("HF_API_KEY", "")})
-              ],
-              volumes={"/root/data": model_vol})
+@app.function(
+    gpu="L40S:1",
+    image=image,
+    timeout=900,
+    secrets=[modal.Secret.from_dict({"HF_API_KEY": os.environ.get("HF_API_KEY", "")})],
+    volumes={"/root/data": model_vol},
+)
 def run_transformer_tests():
     run_test(
         "export HF_HOME='/root/data/.cache' && hf auth login --token $HF_API_KEY && pytest ./fastvideo/tests/transformers -vs"
     )
 
 
-@app.function(gpu="L40S:4",
-              image=image,
-              timeout=900,
-              secrets=[
-                  modal.Secret.from_dict(
-                      {"WANDB_API_KEY": os.environ.get("WANDB_API_KEY", "")})
-              ],
-              volumes={"/root/data": model_vol})
+@app.function(
+    gpu="L40S:4",
+    image=image,
+    timeout=900,
+    secrets=[modal.Secret.from_dict({"WANDB_API_KEY": os.environ.get("WANDB_API_KEY", "")})],
+    volumes={"/root/data": model_vol},
+)
 def run_training_tests():
     run_test(
         "export HF_HOME='/root/data/.cache' && wandb login $WANDB_API_KEY && pytest ./fastvideo/tests/training/Vanilla -srP"
     )
 
 
-@app.function(gpu="L40S:2",
-              image=image,
-              timeout=900,
-              secrets=[
-                  modal.Secret.from_dict(
-                      {"WANDB_API_KEY": os.environ.get("WANDB_API_KEY", "")})
-              ],
-              volumes={"/root/data": model_vol})
+@app.function(
+    gpu="L40S:2",
+    image=image,
+    timeout=900,
+    secrets=[modal.Secret.from_dict({"WANDB_API_KEY": os.environ.get("WANDB_API_KEY", "")})],
+    volumes={"/root/data": model_vol},
+)
 def run_training_lora_tests():
     run_test(
         "export HF_HOME='/root/data/.cache' && wandb login $WANDB_API_KEY && pytest ./fastvideo/tests/training/lora/test_lora_training.py -srP"
     )
 
 
-@app.function(gpu="H100:2",
-              image=image,
-              timeout=900,
-              secrets=[
-                  modal.Secret.from_dict(
-                      {"WANDB_API_KEY": os.environ.get("WANDB_API_KEY", "")})
-              ])
+@app.function(
+    gpu="H100:2",
+    image=image,
+    timeout=900,
+    secrets=[modal.Secret.from_dict({"WANDB_API_KEY": os.environ.get("WANDB_API_KEY", "")})],
+)
 def run_training_tests_VSA():
-    run_test(
-        "wandb login $WANDB_API_KEY && pytest ./fastvideo/tests/training/VSA -srP"
-    )
+    run_test("wandb login $WANDB_API_KEY && pytest ./fastvideo/tests/training/VSA -srP")
 
 
 @app.function(gpu="H100:1", image=image, timeout=900)
@@ -174,33 +160,27 @@ def run_kernel_tests():
 
 @app.function(gpu="L40S:1", image=image, timeout=900)
 def run_inference_tests_vmoba():
-    run_test('python fastvideo/tests/inference/vmoba/test_vmoba_inference.py')
+    run_test("python fastvideo/tests/inference/vmoba/test_vmoba_inference.py")
 
 
 @app.function(gpu="L40S:1", image=image, timeout=1200)
 def run_inference_lora_tests():
-    run_test(
-        "pytest ./fastvideo/tests/inference/lora/test_lora_inference_similarity.py -vs"
-    )
+    run_test("pytest ./fastvideo/tests/inference/lora/test_lora_inference_similarity.py -vs")
 
 
 @app.function(gpu="L40S:2", image=image, timeout=900)
 def run_distill_dmd_tests():
-    run_test(
-        "pytest ./fastvideo/tests/training/distill/test_distill_dmd.py -vs")
+    run_test("pytest ./fastvideo/tests/training/distill/test_distill_dmd.py -vs")
 
 
-@app.function(gpu="L40S:2",
-              image=image,
-              timeout=900,
-              secrets=[
-                  modal.Secret.from_dict(
-                      {"WANDB_API_KEY": os.environ.get("WANDB_API_KEY", "")})
-              ])
+@app.function(
+    gpu="L40S:2",
+    image=image,
+    timeout=900,
+    secrets=[modal.Secret.from_dict({"WANDB_API_KEY": os.environ.get("WANDB_API_KEY", "")})],
+)
 def run_self_forcing_tests():
-    run_test(
-        "wandb login $WANDB_API_KEY && pytest ./fastvideo/tests/training/self-forcing/test_self_forcing.py -vs"
-    )
+    run_test("wandb login $WANDB_API_KEY && pytest ./fastvideo/tests/training/self-forcing/test_self_forcing.py -vs")
 
 
 @app.function(gpu="L40S:1", image=image, timeout=900)
@@ -210,41 +190,36 @@ def run_unit_test():
     )
 
 
-@app.function(gpu="L40S:1",
-              image=image,
-              timeout=3600,
-              secrets=[
-                  modal.Secret.from_dict(
-                      {"HF_API_KEY": os.environ.get("HF_API_KEY", "")})
-              ])
+@app.function(
+    gpu="L40S:1",
+    image=image,
+    timeout=3600,
+    secrets=[modal.Secret.from_dict({"HF_API_KEY": os.environ.get("HF_API_KEY", "")})],
+)
 def run_lora_extraction_tests():
-    run_test(
-        "hf auth login --token $HF_API_KEY && pytest ./fastvideo/tests/lora_extraction/test_lora_extraction.py"
-    )
+    run_test("hf auth login --token $HF_API_KEY && pytest ./fastvideo/tests/lora_extraction/test_lora_extraction.py")
 
 
-@app.function(gpu="L40S:2",
-              image=image,
-              timeout=1800,
-              secrets=[
-                  modal.Secret.from_dict(
-                      {"HF_API_KEY": os.environ.get("HF_API_KEY", "")})
-              ],
-              volumes={"/root/data": model_vol})
+@app.function(
+    gpu="L40S:2",
+    image=image,
+    timeout=1800,
+    secrets=[modal.Secret.from_dict({"HF_API_KEY": os.environ.get("HF_API_KEY", "")})],
+    volumes={"/root/data": model_vol},
+)
 def run_performance_tests():
     run_test(
         "export HF_HOME='/root/data/.cache' && hf auth login --token $HF_API_KEY && pytest ./fastvideo/tests/performance -vs"
     )
 
 
-@app.function(gpu="L40S:1",
-              image=image,
-              timeout=1800,
-              secrets=[
-                  modal.Secret.from_dict(
-                      {"HF_API_KEY": os.environ.get("HF_API_KEY", "")})
-              ],
-              volumes={"/root/data": model_vol})
+@app.function(
+    gpu="L40S:1",
+    image=image,
+    timeout=1800,
+    secrets=[modal.Secret.from_dict({"HF_API_KEY": os.environ.get("HF_API_KEY", "")})],
+    volumes={"/root/data": model_vol},
+)
 def run_api_server_tests():
     run_test(
         "export HF_HOME='/root/data/.cache' && hf auth login --token $HF_API_KEY && pytest ./fastvideo/tests/entrypoints/test_openai_api_integration.py -vs"

fastvideo/tests/ssim/inference_similarity_utils.py

@@ -79,13 +79,8 @@ def _assert_similarity(
     attention_backend_name: str,
 ) -> None:
     if not os.path.exists(reference_folder):
-        logger.error("Reference folder missing: %s", reference_folder)
-        error_msg = (
-            f"Reference video folder does not exist: {reference_folder}\n"
-            f"To download reference videos, run:\n"
-            f"  python fastvideo/tests/ssim/reference_videos_cli.py download"
-        )
-        raise FileNotFoundError(error_msg)
+        logger.error("Reference folder missing")
+        raise FileNotFoundError(f"Reference video folder does not exist: {reference_folder}")
 
     try:
         reference_video_path = _find_reference_video(reference_folder, prompt)

fastvideo/tests/ssim/reference_videos_cli.py

@@ -51,7 +51,11 @@ def _iter_video_files(root: Path) -> Iterable[Path]:
 def _discover_reference_dirs(base_dir: Path) -> list[Path]:
     if not base_dir.exists():
         return []
-    dirs = [p for p in base_dir.iterdir() if p.is_dir() and p.name.endswith("_reference_videos")]
+    dirs = [
+        p
+        for p in base_dir.iterdir()
+        if p.is_dir() and p.name.endswith("_reference_videos")
+    ]
     return sorted(dirs)
 
 
@@ -102,7 +106,10 @@ def _resolve_device_reference_folder(
 ) -> str:
     if explicit_device_folder and explicit_device_folder.strip():
         return explicit_device_folder.strip()
-    if reference_dir is not None and reference_dir.name.endswith("_reference_videos"):
+    if (
+        reference_dir is not None
+        and reference_dir.name.endswith("_reference_videos")
+    ):
         return reference_dir.name
     return DEFAULT_DEVICE_REFERENCE_FOLDER
 
@@ -134,11 +141,6 @@ def _discover_reference_dirs_for_tier(
 
 
 def _has_local_reference_videos(base_dir: Path, quality_tier: str) -> bool:
-    # Check for completion marker file first
-    marker_path = base_dir / REFERENCE_VIDEOS_DIRNAME / quality_tier / f".download_complete_{quality_tier}"
-    if not marker_path.exists():
-        return False
-    # Also verify at least one .mp4 exists
     for ref_dir in _discover_reference_dirs_for_tier(base_dir, quality_tier):
         for _ in _iter_video_files(ref_dir):
             return True
@@ -159,7 +161,8 @@ def _load_hf_sdk():
         from huggingface_hub import HfApi, snapshot_download
     except ImportError as exc:
         raise RuntimeError(
-            "huggingface_hub is required for download/upload.\nInstall with: pip install huggingface_hub"
+            "huggingface_hub is required for download/upload.\n"
+            "Install with: pip install huggingface_hub"
         ) from exc
     return HfApi, snapshot_download
 
@@ -207,7 +210,9 @@ def download_reference_videos(
     for quality_tier in tiers:
         tier_prefix = f"{REFERENCE_VIDEOS_DIRNAME}/{quality_tier}"
         if device_folders:
-            allow_patterns.extend(f"{tier_prefix}/{folder}/**" for folder in device_folders)
+            allow_patterns.extend(
+                f"{tier_prefix}/{folder}/**" for folder in device_folders
+            )
         else:
             allow_patterns.append(f"{tier_prefix}/*_reference_videos/**")
 
@@ -237,12 +242,9 @@ def download_reference_videos(
         )
         if copied_roots == 0:
             raise RuntimeError(
-                "HF download completed but no reference_videos content was found in downloaded artifacts."
+                "HF download completed but no reference_videos content "
+                "was found in downloaded artifacts."
             )
-        for quality_tier in tiers:
-            marker_path = local_dir / REFERENCE_VIDEOS_DIRNAME / quality_tier / f".download_complete_{quality_tier}"
-            marker_path.parent.mkdir(parents=True, exist_ok=True)
-            marker_path.touch()
 
 
 def upload_reference_videos(
@@ -264,8 +266,12 @@ def upload_reference_videos(
 
     for quality_tier, reference_dir in reference_dirs_by_tier:
         if not reference_dir.exists():
-            raise FileNotFoundError(f"Reference directory not found: {reference_dir}")
-        path_in_repo = f"{REFERENCE_VIDEOS_DIRNAME}/{quality_tier}/{reference_dir.name}"
+            raise FileNotFoundError(
+                f"Reference directory not found: {reference_dir}"
+            )
+        path_in_repo = (
+            f"{REFERENCE_VIDEOS_DIRNAME}/{quality_tier}/{reference_dir.name}"
+        )
         print(f"Uploading {reference_dir.name} ({quality_tier}) to {repo_id} ...")
         api.upload_folder(
             repo_id=repo_id,
@@ -288,14 +294,20 @@ def _resolve_upload_reference_dirs(
 
     if explicit_reference_dirs:
         if len(tiers) != 1:
-            raise RuntimeError("--reference-dir requires a single --quality-tier (default or full_quality).")
+            raise RuntimeError(
+                "--reference-dir requires a single --quality-tier "
+                "(default or full_quality)."
+            )
         resolved_tier = tiers[0]
         return [(resolved_tier, Path(p).resolve()) for p in explicit_reference_dirs]
 
     resolved: list[tuple[str, Path]] = []
     for tier in tiers:
         for reference_dir in _discover_reference_dirs_for_tier(base_dir, tier):
-            if selected_device_folders and reference_dir.name not in selected_device_folders:
+            if (
+                selected_device_folders
+                and reference_dir.name not in selected_device_folders
+            ):
                 continue
             resolved.append((tier, reference_dir.resolve()))
     return resolved
@@ -315,38 +327,32 @@ def ensure_reference_videos_available(
     lock_path = target_dir / ".reference_videos_download.lock"
     with _exclusive_download_lock(lock_path):
         if _has_local_reference_videos(target_dir, quality_tier):
-            print(f"Reference videos ({quality_tier}) already available at {target_dir}")
             return False
 
         resolved_repo_id = repo_id or _default_repo_id()
         resolved_repo_type = repo_type or _default_repo_type()
         if not resolved_repo_id:
             raise RuntimeError(
-                f"No local reference videos found and no HF repo configured.\nSet {HF_REPO_ENV_KEY} or pass --repo-id."
+                "No local reference videos found and no HF repo configured.\n"
+                f"Set {HF_REPO_ENV_KEY} or pass --repo-id."
             )
 
-        print(f"Repo ID: {resolved_repo_id}")
-        print(f"Quality tier: {quality_tier}")
-        print(f"No local {quality_tier} reference videos found under {target_dir}. Starting download...")
-        try:
-            download_reference_videos(
-                repo_id=resolved_repo_id,
-                repo_type=resolved_repo_type,
-                local_dir=target_dir,
-                quality_tiers=[quality_tier],
-            )
-            print(f"Download completed for {quality_tier} reference videos.")
-        except Exception as exc:
-            print(f"ERROR: Failed to download {quality_tier} reference videos from {resolved_repo_id}.")
-            print(
-                f"Suggested command to retry: "
-                f"python fastvideo/tests/ssim/reference_videos_cli.py download "
-                f"--quality-tier {quality_tier}"
-            )
-            raise
+        print(
+            f"No local {quality_tier} reference videos found under {target_dir}. "
+            f"Downloading from HF repo {resolved_repo_id} ..."
+        )
+        download_reference_videos(
+            repo_id=resolved_repo_id,
+            repo_type=resolved_repo_type,
+            local_dir=target_dir,
+            quality_tiers=[quality_tier],
+        )
 
         if not _has_local_reference_videos(target_dir, quality_tier):
-            raise RuntimeError(f"HF download completed but no {quality_tier} *_reference_videos content found.")
+            raise RuntimeError(
+                f"HF download completed but no {quality_tier} *_reference_videos "
+                "content found."
+            )
         return True
 
 
@@ -391,12 +397,18 @@ def _build_parser() -> argparse.ArgumentParser:
         "--generated-dir",
         type=Path,
         default=None,
-        help=("Source generated directory. Default: <ssim_dir>/generated_videos/<quality-tier>/<device-folder>"),
+        help=(
+            "Source generated directory. "
+            "Default: <ssim_dir>/generated_videos/<quality-tier>/<device-folder>"
+        ),
     )
     copy_parser.add_argument(
         "--device-folder",
         default=None,
-        help=("GPU folder name (e.g., H200_reference_videos) used to build default generated/reference paths."),
+        help=(
+            "GPU folder name (e.g., H200_reference_videos) used to build "
+            "default generated/reference paths."
+        ),
     )
     copy_parser.add_argument(
         "--reference-dir",
@@ -488,13 +500,19 @@ def _build_parser() -> argparse.ArgumentParser:
         "--device-folder",
         action="append",
         default=[],
-        help=("Specific GPU reference folder to upload (repeatable), e.g. H200_reference_videos."),
+        help=(
+            "Specific GPU reference folder to upload (repeatable), "
+            "e.g. H200_reference_videos."
+        ),
     )
     upload_parser.add_argument(
         "--base-dir",
         type=Path,
         default=_ssim_dir(),
-        help=("Base SSIM directory that contains reference_videos/<quality-tier>/<GPU>_reference_videos."),
+        help=(
+            "Base SSIM directory that contains "
+            "reference_videos/<quality-tier>/<GPU>_reference_videos."
+        ),
     )
     upload_parser.add_argument(
         "--private",
@@ -520,7 +538,10 @@ def _build_parser() -> argparse.ArgumentParser:
         "--local-dir",
         type=Path,
         default=_ssim_dir(),
-        help=("Local SSIM directory that should contain reference_videos/<quality-tier>/<GPU>_reference_videos."),
+        help=(
+            "Local SSIM directory that should contain "
+            "reference_videos/<quality-tier>/<GPU>_reference_videos."
+        ),
     )
     ensure_parser.add_argument(
         "--quality-tier",
@@ -556,20 +577,21 @@ def main(argv: Sequence[str] | None = None) -> int:
         reference_dir = (
             args.reference_dir
             if args.reference_dir is not None
-            else (_ssim_dir() / REFERENCE_VIDEOS_DIRNAME / args.quality_tier / device_folder)
+            else (
+                _ssim_dir()
+                / REFERENCE_VIDEOS_DIRNAME
+                / args.quality_tier
+                / device_folder
+            )
         )
         copied = copy_generated_to_reference(
             generated_dir=generated_dir,
             reference_dir=reference_dir,
             dry_run=args.dry_run,
         )
-        print(f"Done. {'Would copy' if args.dry_run else 'Copied'} {copied} video files.")
-        if not args.dry_run and copied > 0:
-            marker = (
-                _ssim_dir() / REFERENCE_VIDEOS_DIRNAME / args.quality_tier / f".download_complete_{args.quality_tier}"
-            )
-            marker.parent.mkdir(parents=True, exist_ok=True)
-            marker.touch()
+        print(
+            f"Done. {'Would copy' if args.dry_run else 'Copied'} {copied} video files."
+        )
         return 0
 
     if args.command == "download":
@@ -581,14 +603,19 @@ def main(argv: Sequence[str] | None = None) -> int:
             quality_tiers=quality_tiers,
             device_folders=args.device_folder or None,
         )
-        print("Download complete for quality tier(s): " + ", ".join(quality_tiers) + ".")
+        print(
+            "Download complete for quality tier(s): "
+            + ", ".join(quality_tiers)
+            + "."
+        )
         return 0
 
     if args.command == "upload":
         token = _get_hf_token()
         if token is None:
             raise RuntimeError(
-                "Hugging Face API key is required for upload. Set HF_API_KEY, HUGGINGFACE_HUB_TOKEN, or HF_TOKEN."
+                "Hugging Face API key is required for upload. Set "
+                "HF_API_KEY, HUGGINGFACE_HUB_TOKEN, or HF_TOKEN."
             )
 
         reference_dirs_by_tier = _resolve_upload_reference_dirs(
@@ -599,7 +626,8 @@ def main(argv: Sequence[str] | None = None) -> int:
         )
         if not reference_dirs_by_tier:
             raise RuntimeError(
-                f"No *_reference_videos directories found for the selected quality tier(s) under {args.base_dir}"
+                "No *_reference_videos directories found for the selected "
+                f"quality tier(s) under {args.base_dir}"
             )
 
         upload_reference_videos(
@@ -620,7 +648,10 @@ def main(argv: Sequence[str] | None = None) -> int:
             quality_tier=args.quality_tier,
         )
         if downloaded:
-            print(f"{args.quality_tier} reference videos were missing and have been downloaded.")
+            print(
+                f"{args.quality_tier} reference videos were missing and "
+                "have been downloaded."
+            )
         else:
             print(f"{args.quality_tier} reference videos already exist locally.")
         return 0